<aside> <img src="https://s3-us-west-2.amazonaws.com/secure.notion-static.com/72f61cfe-f296-4060-b340-2df0adb4c596/original_copy.gif" alt="https://s3-us-west-2.amazonaws.com/secure.notion-static.com/72f61cfe-f296-4060-b340-2df0adb4c596/original_copy.gif" width="40px" /> Hello! I’m an M.S. Applied Data Science candidate at the University of Chicago Data Science Institute graduating in December 2026. My work spans machine learning, product analytics, data engineering, geospatial modeling, and human-centered technology.

I am interested in building data products that help people understand complex information, make better decisions, and interact more meaningfully with technology. My background in Media Arts and Design also shapes how I approach data science: with attention to interpretation, communication, and the experience of the people using a system.

</aside>

Featured Work

Predicting the Impact of Development on Biodiversity and Ecosystem Loss

University of Chicago Applied Data Science Capstone – Shutay Consulting March 2026-Present

A large-scale geospatial modeling project designed to estimate how future urban development may affect biodiversity and ecosystem health over a 30-year horizon.

I am co-leading the project’s technical design, research scope, client communication, and modeling methodology. The system combines approximately 20 years of species-occurrence, climate, ecosystem, and human-development data within a shared global spatial framework.

Current Work

Implemented an ISEA3H discrete global grid to align datasets with different geographic structures and resolutions.
Aggregating biodiversity and environmental data across approximately 20 years of spatial and temporal observations.
Investigating observation bias in citizen-science datasets such as GBIF and eBird, where recorded observations reflect human activity as well as underlying species presence.
Implemented country-year rarefaction as a baseline method for reducing unequal sampling intensity.
Evaluating probabilistic methods for representing remaining uncertainty in species observations and model predictions.
Designing a three-part pipeline for forecasting urban expansion, ecosystem condition, and biodiversity risk.
Defining development scenarios representing stable growth, unconstrained growth, and accelerating urban sprawl.

Methods and tools: Python, geospatial data processing, ISEA3H, feature engineering, statistical modeling, time-series analysis, uncertainty modeling, Azure Databricks

Week 10 Capstone Presentation.pdf